Our goal is to develop and optimize assays for the diagnosis and prediction of type 1A diabetes and to discover genetic determinants of the expression of anti-islet autoimrrfunity and subsequent diabetes. These goals are complementary in that detection of anti-islet autoantibodies provides essential phenotypic information for genetic analysis. We developed rapid assays to allow "real-time" analysis of serum from cadaveric donors and analyzed expression of autoantibodies in multiple populations. Findings include: 1) Screening and follow up of autoantibody positive children can preveriit ketoacidosis, 2) Approximately 1/200 U.S. cadaveric donors express multiple anti-islet autoantibodies 3) More than 40% of highest risk class II HLA DR3-DQ2/DR4-DQ8 siblings of patients with type 1 diabetes express anti-islet autoantibodies by age 6 but less than 5% of DR3-DQ2/DR4-DQ8 general population children!by age 6 develop suchautoantibodies ("Relative DR3/4 Paradox"), 4) In the large DPT-1 trials multiple antijislet autoantibodies are highly predictive of progression to diabetes, but there is an important antigen(s)-"X" spch that cytoplasmic ICA plus any single biochemical autoantibody is associated with greatly enhanced progression to diabetes and 7% of ICA positive biochemical autoantibody negative relatives progress to diabetes, and 5)Collaborated in the discovery of a LYP (Lymphoid Tyrosine Phosphatase: PTPN22) autcpimmunity polymorphism. In this proposal, we extend our studies of the affinity and epitope specificity of insulin autoantibodies with our recently improved mlAA assay and adapt a published insulin autoantibody assay that should allow us to engineer insulin analogues to more precisely define epitope recognition, will carry out preliminary studies of the reactivity of ICA positive sera with the pancreas from the GM2/G|D2 Synthase Knockout mouse [Galgt- 1(GM) KO] to initially assess contribution of complex gangliosides to' ICA reactivity of biochemical autoantibody negative sera, and study the "Relative DR3/4 Paradox" by characterizing in detail the major histocompatibility complex (MHC) using a high density MHC snp map (as well as analysis of LYP polymorphisms in combination with insulin, MIC-A and CTLA-4, andother polymorphisms).